CN101116280A - Synchronisation device and device for generation of a synchronisation signal - Google Patents

Abstract

A synchronization device for determining a position of a synchronization signal in a receive signal, the synchronization signal being based on a coarse synchronization signal and a fine synchronization signal, includes a signal processing means configured to determine, based on the coarse synchronization signal, a section of the receive signal in which is located the fine synchronization signal, and to determine in the section of the receive signal, based on the fine synchronization signal, the position of the synchronization signal in the receive signal for a synchronization.

Description

Synchronization apparatus and apparatus for generating synchronization signal

Technical Field

The invention relates to synchronizing a receiver and a transmitter in a digital transmission system.

Background

In a digital transmission system, the information bits to be transmitted are combined into information blocks, each information block corresponding to a frame structure. In addition to the information bits, each information block also includes a number of additional information bits needed to enable data transmission.

The receiver can detect the transmitted information and, in addition to retrieving the samples, the receiver must also determine the point in time at which the corresponding information block was received. In other words, the receiver must perform block synchronization in order to synchronize the receiver and the transmitter with respect to each other.

To achieve synchronization, for example, the transmitter sends out a synchronization signal known to the receiver. In a receiver, a received signal including a synchronization signal is processed to detect a position of the synchronization signal in the received signal and/or to detect a time at which the synchronization signal is generated. For this purpose, for example, a cross-correlation between the received signal and a copy of the synchronization signal has to be performed in order to find the synchronization signal in the received signal. Since a wideband synchronization signal has to be transmitted in order to determine the position accurately, determining the position involves considerable computational complexity, since the wideband signal always needs to be processed with a high sampling clock.

For example, if synchronization is to be done with a very wideband digital signal (RF wideband B), i.e. if a fixed known synchronization signal (preamble) is to be found in the continuous received signal, or if a preamble is to be found in the signal part (signal burst), the received signal has to be sampled and processed at least with the bandwidth B to find the synchronization signal, according to the sampling law. For example, if the transmitter or receiver has just been turned on, there is no information yet about where the synchronization signal (preamble) may be.

For this reason, the entire received signal must be sampled and processed over a relatively long period of time with a high bandwidth B. If sampling is done with a bandwidth of 77MHz to 100MHz, one hundred million tests per second must be performed with respect to preamble start, i.e., with respect to the start of the preamble in the received signal, in order to find the preamble without a priori information. For example, if cross-correlation is performed once in each test, each test involves a very large number of computational operations, which requires considerable system resources. For example, due to this high consumption, a direct or blind (online) search for preambles in the wideband signal is not possible.

Disclosure of Invention

It is an object of the invention to provide a synchronization concept that reduces the consumption.

This object is achieved by a synchronization device according to claim 1, a device for synchronizing a receiver and a transmitter according to claim 24, a device for generating a synchronization signal according to claim 25, a synchronization method according to claim 42, a method for synchronizing a receiver and a transmitter according to claim 43, a method for generating a synchronization signal according to claim 44 or a computer program according to claim 45.

The present invention is based on the following findings: if the synchronization is performed in two stages, the synchronization of the receiver and the transmitter can be performed in an efficient manner with reduced consumption. According to the present invention, coarse synchronization is initially performed based on a narrow-band coarse synchronization signal, which is a part of a synchronization signal. According to the present invention, a part of the received signal including the synchronization signal is initially detected in this manner. Since the coarse synchronization signal is narrowband, coarse synchronization can be performed in a cost-effective manner at a low processing rate. In a second stage, a wideband fine synchronization signal, also included in the synchronization signal, is searched in the detected portion of the received signal to accurately determine the location of the fine synchronization signal in a portion of the received signal to accurately locate the synchronization signal in the received signal. Since the fine synchronization signal is wider than the coarse synchronization signal in bandwidth, accurate determination of the position of the synchronization signal in the received signal is performed at a higher processing rate than the coarse synchronization signal.

The synchronization concept of the present invention is based on the following facts: the synchronization signal transmitted by the transmitter is based on the coarse synchronization signal and the fine synchronization signal. Preferably, the synchronization signal comprises a coarse synchronization signal and a fine synchronization signal, the fine synchronization signal is preferably transmitted temporally after the coarse synchronization signal, and the bandwidth of the fine synchronization signal is wider than the bandwidth of the coarse synchronization signal.

In order to determine the position of the synchronization signal in the received signal for synchronization, the invention provides a signal processing device which is configured to determine a part of the received signal in which the fine synchronization signal is located from the coarse synchronization signal and to determine the position of the synchronization signal in the received signal in the part of the received signal for synchronization from the fine synchronization signal.

Preferably, the signal processing means is arranged to detect a coarse synchronization signal in the received signal to determine a portion of the received signal. The partial received signal is determined by, for example, the detection time (detection time) of the coarse synchronization signal. In order to determine the position of the synchronization signal in the received signal, the signal processing means are arranged for detecting a fine synchronization signal in a part of the received signal determined approximately from the coarse synchronization signal.

According to the present invention, since the bandwidth of the coarse synchronization signal is smaller than that of the fine synchronization signal, the detection of the coarse synchronization signal can be performed at a low processing rate to save signal processing resources. However, the detection of the fine synchronization signal is performed at a higher processing rate, and thus processing resources can be allocated in a targeted manner. To this end, the signal processing means are arranged for performing a detection of a coarse synchronization signal in the received signal at a first sampling rate and a detection of a fine synchronization signal at a second sampling rate, the second sampling rate being higher than the first sampling rate. The term "sampling rate" refers to the term "processing rate" and will be used below to indicate the number of operations per time unit (i.e., clock rate).

According to the sampling law, since the sampling rate is predetermined by the signal bandwidth, the search for a coarse synchronization signal, which is preferably narrowband, can be performed at a low sampling clock (processing clock). In contrast, the search for a fine synchronization signal is performed at a higher processing clock associated with a higher sampling rate. Therefore, in the case where the bandwidth ratio of the coarse synchronization signal to the fine synchronization signal is large, the reduction in consumption achieved during the first synchronization stage (coarse synchronization) is significant.

According to the invention, the synchronization process is also accelerated, since in the coarse synchronization phase, preferably only the part in which the fine synchronization signal is located is approximately determined. In order to determine the portion of the received signal it is sufficient that the signal processing means only detects the presence of a coarse synchronization signal.

Preferably, the signal processing means is arranged to perform a correlation (e.g. cross-correlation) between the received signal and a signal related to the coarse synchronization signal to detect the coarse synchronization signal in the received signal. The correlation may be performed continuously at a low sampling rate, thus searching the received signal at reduced consumption to approximately determine the portion of the received signal where the fine synchronization signal is located.

In order to determine the position of the fine synchronization signal in the partial received signal, i.e. the position of the synchronization signal in the received signal, the signal processing means are preferably configured to perform a correlation between the partial received signal and a signal related to the fine synchronization signal. According to the present invention, the correlation is performed at a high sampling rate and a high processing clock so as to accurately detect the position of the synchronization signal.

For example, the signal associated with the coarse synchronization signal may be a copy of the coarse synchronization signal known to the receiver. And so on, the signal associated with the fine synchronization signal may be a copy of the fine synchronization signal known to the receiver.

As described above, the synchronization signal includes a coarse synchronization signal (hereinafter also referred to as an acquisition burst or an a burst) and a fine synchronization signal (also referred to as a tracking burst or a T burst). For example, the synchronization signal (preamble) includes two parts, and therefore, according to the present invention, a search for division is performed. The first part of the preamble exhibits a relatively low bandwidth BA, enabling an on-line search, i.e. a search that is continued over a relatively long period of time until a fine synchronization signal is found. To this end, for example, the corresponding sub-frequency range of the bandwidth BA is filtered out of the received signal of the bandwidth B and sampled at a relatively low sampling rate (of the order of BA). Other data processing and searching must also be done with clock BA. Due to this a-burst, the approximate position of another preamble (fine synchronization signal) can be determined. However, since the bandwidth BA is low, the position accuracy is very low. In contrast, for example, the fine synchronization signal exhibits a full bandwidth B. A search window is opened at the expected position of the T burst (part of the received signal) of the received signal predicted by the a burst. With this window, the received signal of full bandwidth B is sampled and processed with clock B. Due to the limitation of one window, the position of the T-burst can be determined with high accuracy. To this end, the correlation is calculated as described above as an example. For example, with a bandwidth BA =6MHz and a bandwidth B =77MHz, a consumption reduction by a factor of 16 can be achieved with respect to the processing with clock B.

Drawings

Further preferred embodiments of the invention will be explained in more detail below with reference to the drawings, in which:

FIG. 1 is a basic block diagram of the synchronization apparatus of the present invention for determining the position of a synchronization signal in a received signal;

FIG. 2 is a block diagram of a synchronization apparatus according to another embodiment;

FIG. 3 shows a time relationship;

FIG. 4 shows, as an example, four time signals;

FIG. 5 is a calculation specification illustration;

FIG. 6 is a flow chart of the signal processing of the present invention;

FIG. 7 is an example of a pre-trigger;

FIG. 8 is a calculation specification description;

fig. 9 shows a basic construction for acquisition of the inventive receiver;

FIG. 10 is a related specification description;

FIG. 11 is a constellation diagram;

FIG. 12 is a transfer function of a filter;

fig. 13a shows the distribution of complex signals in 9 different frequency bands;

FIG. 13b is a frequency table;

fig. 14 shows an a-burst;

FIG. 15 shows distances in burst multiplexing;

FIG. 16 is a block diagram of a systematic encoder for CRC-12 codes;

FIG. 17 is a block diagram of a time-limited recursive systematic convolutional encoder; and

fig. 18 shows the generation of a T-burst.

Detailed Description

The synchronization device comprises a first detection means 101, an input and an output of the first detection means 101 being connected to a second detection means 103. The second sensing device has an input coupled to the input of the first sensing device and has an output.

According to the invention, a received signal comprising a synchronization signal consisting of a coarse synchronization signal 105 and a fine synchronization signal 107 may be applied to the input of the first detection means 101, as shown in fig. 1.

The first detection means 101 are arranged for detecting a coarse synchronization signal in the received signal for detecting a part of the received signal where the fine synchronization signal is located. Upon detection of the coarse synchronization signal, the first detection means 101 provide via output information about a part of the received signal where the fine synchronization signal is located to the second detection means 103. The second detection means 103 is arranged to detect the position of the fine synchronization signal in the partial reception signal and to output information about the synchronization signal position specified by the position of the fine synchronization signal in the partial reception signal. The first detection means are arranged for detecting the coarse synchronization signal at a first sampling rate, i.e. at a first clock. In contrast, the second detection means 103 are configured for detecting the position of the fine synchronization signal at the second sampling rate, i.e. at the second clock. The first sampling rate is lower than the second sampling rate (the first clock is lower than the second clock).

According to the present invention, the bandwidth of the coarse synchronization signal is smaller than the bandwidth of the fine synchronization signal. For detecting the coarse synchronization signal, the first detection means 101 comprises, for example: and the filter is used for filtering the received signal to filter out the coarse synchronization signal from the received signal. If the coarse synchronization signal is a band pass signal having a center frequency, the coarse synchronization signal occupies a predetermined frequency range defined by the center frequency and a bandwidth of the coarse synchronization signal. The received signal components occupying the predetermined frequency range are thus filtered out by means of a filter, which may be a band-pass filter. If the coarse synchronization signal has been sent out, its received version is included in the received signal component after filtering.

If the coarse synchronization signal is a band pass signal, the first detection means comprises, for example: a down converter (down converter) for down converting a received signal component, which is also a band pass signal, to obtain a received signal of the baseband, which received signal component is provided for detecting the coarse synchronization signal in the baseband.

However, the coarse synchronization signal may already be a baseband signal, so the predetermined frequency range is in baseband. In this case, the filter is a low-pass filter for filtering out the received signal component in the baseband.

If the received signal is an analog signal, the first detection means 101 may further comprise sampling means for performing an analog/digital conversion of the received signal component at the first sampling rate.

According to another aspect, the signal processing apparatus of the present invention includes, for example: controllable sampling means configured to sample the received signal at a first sampling rate when the coarse synchronization signal has not been detected by the first detection means 101 and at a second sampling rate when the coarse synchronization signal has been detected by the first detection means 101. To this end, the synchronization device may further comprise control means for controlling the sampling means in response to a detection signal indicating that the coarse synchronization signal is detected and that it can be output by the first detection means 101, so as to set a higher sampling rate, for example a second sampling rate.

According to another aspect of the invention, the received signal may already be a digital signal sampled at the second sampling rate. For the detection of the coarse synchronization signal at the first sampling rate, the first detection device 101 may further comprise a sampling rate converter configured to down-sample the received signal to obtain a received signal at the first sampling rate.

According to another aspect of the invention, for detecting the coarse synchronization signal in the received signal, or in a received signal component being part of the received signal, the first detection means 101 comprises a detector, which is for example arranged to perform a correlation between the received signal component and a signal related to the coarse synchronization signal to a sufficient extent. For example, the signal associated with the coarse synchronization signal may be a baseband or bandpass copy of the coarse synchronization signal.

To indicate the detection of the coarse synchronization signal, for example, a detector comprised in the first detection means 101 is configured to output a detection signal, for example comprising information about a part of the received signal where the fine synchronization signal is located. For example, the detector may be configured to output a detection signal when the correlation value exceeds a detection threshold. For example, the correlation value may be a cross-correlation coefficient at a zero point. Furthermore, the correlation values may be normalized to the maximum correlation value, such that the detection threshold may be a relative magnitude, preferably with a relative (i.e. normalized) correlation value of e.g. 10%. In other words, the power normalization can be performed in such a way that, ideally, the received signal is undistorted compared to the transmitted signal (received signal = factor x transmitted signal), achieving a maximum of the normalized correlation value. In non-ideal cases, i.e. in the case of distortion of the received signal, the normalized correlation value is smaller than the maximum value.

For example, the detection signal provided by the first detection means 101 may indicate a detection time at which the coarse synchronization signal is detected, the detection time indicating a start of a portion of the received signal at which the fine synchronization signal is located. The second detection means 103 is arranged to receive the detection signal and, in response to the detection signal, detect the position of the fine synchronization signal in a part of the received signal specified by the detection time.

According to another aspect of the invention the first detection means 101 are arranged for activating the second detection means 103 such that the second detection means 103 start detecting the position of the fine synchronization signal after the detection time, thus detecting the position of the synchronization signal in the received signal.

Fig. 2 shows a block diagram of a synchronization device according to another embodiment.

The synchronization device comprises first detection means 101 and second detection means 103. For illustrative reasons, the function of the synchronization device is separated by the longitudinal score line in fig. 2 in order to clarify the two-step method of the present invention. The function depicted to the left of the vertical scribe is essentially responsible for the acquisition (acquisition algorithm), i.e. for detecting the coarse synchronization signal in the received signal. In contrast, the function illustrated on the right of the longitudinal dashed line is responsible for precisely determining the position of the fine synchronization signal, i.e. the position of the synchronization signal in a part of the received signal (tracking algorithm).

The second detection means 103 shown in fig. 2 comprise delay means 201 for delaying the received signal, which delay means are arranged for compensating the detection delay of the first detection means 101. The delay means 210 may be, for example: a first-in-first-out (FIFO) memory having a memory depth to achieve the desired delay.

The delay means 201 comprises an output connected to a detector 203 comprised in the second detection means 103. The detector 203 is configured to detect a position of the fine synchronization signal in a portion of the received signal. For example, the detector 203 is configured to determine a correlation between the received signal and a signal related to the fine synchronization signal to detect a position of the fine synchronization signal.

For example, the signal associated with the fine synchronization signal may be a copy of the baseband or bandpass signal.

Furthermore, in the embodiment shown in fig. 2, the second detection means 103 comprise a first correlator 205. The first correlator 205 includes: an input connected to an output of the delay means 201, another input connected to an output of the first detection means 101, and a second output connected to an input of the second correlator 207 or the interpolation means 207.

For example, the first and second correlators 205, 207 may be included in a detector as described in connection with the embodiment illustrated in FIG. 1.

For example, the received signals applied at the input of the first detection means 101 and at the input of the second detection means 103 are baseband signals provided by the antenna elements and down-converted to the (complex) baseband and thus present in the I and Q components. For example, the received signal has been sampled with a sampling clock B _ cloc, so the received signal has a digital form.

The first detection means 101 are arranged for performing acquisition, i.e. for performing a continuous search of the a-burst (coarse synchronization signal) for synchronization with the transmitter. For example, if the received signal comprises synchronization signals comprised in a plurality of transmitters, each synchronization signal comprising a coarse synchronization signal occupying a different frequency band, the filter comprised in the first detection means 101 is tunable to filter out received signal components occupying the frequency range in which the coarse synchronization signal to be detected is located.

The first correlator 205 is configured for receiving a signal related to a part of the received signal (window position) via said further input such that the computationally intensive part of the tracking algorithm, i.e. the position of the fine synchronization signal, is performed only in the part of the received signal (window). The first correlator 205 is configured to perform a search of the T-burst with a grid (raster) B _ sample. To this end, the first correlator 205 is configured to determine a correlation between a portion of the received signal and a signal related to the fine synchronization signal to find the fine synchronization signal (T-burst). The second correlator 207 is configured to perform a fine correlation with the oversampling clock. The second correlator 207 searches the T-burst in a very fine grid using the correlation values determined by the first correlator 205.

The acquisition algorithm performed by the first detection means 107 uses the coarse synchronization signal (a burst) in the received signal to predict the approximate position of the T burst in the received signal.

The received signal is delayed in a calculated manner by means of the FIFO 201 before being input to the tracking algorithm. The tracking algorithm cuts out a window of the received signal to search for the T-burst due to the location prediction of the acquisition algorithm.

According to the invention, the search is performed in two steps. In a first step, the correlation value of the T-burst with the grid B sample is determined in a prediction window. The correlation is performed by a first correlator 205.

According to another aspect of the invention, the second correlator 207 is configured for determining a correlation based on the partial correlation to substantially reduce the effect of the frequency offset on the correlation value. The frequency offset is the frequency difference between the transmitter and the receiver. To this end, for example, the first correlator 205 comprised in the above-mentioned detector is configured to determine a first partial correlation between a first subset of received signal values in the partial received signal and a first subset of signal values related to the fine synchronization signal, and to determine a second partial correlation between a second subset of received signal values in the partial received signal and a second subset of signal values related to the fine synchronization signal. The first subset and the second subset differ by at least one value. Preferably, the first subset and the second subset are different parts of the respective signal. The correlation is determined by the overlap of the first and second partial correlations. To remove the phase offset, the detector and/or the first correlator 205 may be configured to detect a phase relationship between corresponding values of the first and second partial correlations, the phase relationship comprising a frequency offset between the transmitter and the receiver. The respective values of the first and second partial correlations are partial correlation values generated at the same positions of the respective partial correlations. Based on the determined phase relationship, the detector and/or the first correlator 205 is configured to cancel the phase relationship by weighting the first partial correlation or weighting the second partial correlation to reduce the effect of the frequency offset on the correlation value. The respective correlation values are estimated, for example, by a reverse phase relationship.

According to another aspect of the invention, the detector comprised in the second detection means 103 comprises an interpolator for interpolating between the correlated values to obtain a fine correlation and detecting the position of the fine synchronization signal with a higher accuracy based on the fine correlation. In the embodiment shown in fig. 2, the interpolation is performed, for example, by a second interpolator 207, which second interpolator 207 may be configured to perform oversampling, wherein zeros are inserted between the related values to perform the interpolation. For interpolation, an interpolator may be included in the second correlator 207 and may be configured, for example, as an interpolation filter.

According to another aspect of the present invention, the fine synchronization signal may further include information transmitted together with the synchronization information. For example, the information may be encoded by the phase relationship between successive values of the fine synchronization signal. For example, the phase relationship may be a 180 ° phase jump occurring at a predetermined position in the fine synchronization signal, so that, for example, the latter half of the fine synchronization signal is rotated by an additional 180 °, for example multiplied by-1. The second detection means 103 are arranged for detecting the information by detecting the phase relation. This preferably includes detecting phase jumps so that information can be retrieved from the phase by phase decoding.

According to another aspect, the second detection means 103 is configured for deriving the quality value of the reception quality from the correlation. Here, for example, the correlation can be used to derive the channel attenuation which has an effect on the quality of the received signal. Furthermore, for example, with correlation, the power of additive spurious signals (e.g., noise) can be detected that also have an effect on the quality of the received signal.

The correlation value, the estimated phase relationship corresponding to the estimated channel bits, and the quality value are determined by the second correlator 205. Furthermore, the first correlator 205 may be configured to output a signaling information (signaling flag). For example, this information may be passed to a software module that interpolates the correlation values in a finer grid than B _ sample (fine correlation), which may find a highly accurate estimate of the arrival time of the synchronization signal. However, the interpolation may also be determined in the second correlator 207 (or in the interpolation device 207). Further, other quality values may be determined based on the correlation values.

According to another embodiment, the second detection means are arranged for outputting a position signal indicative of the position of the fine synchronization signal in the portion of the received signal, to indicate the position of the synchronization signal in the received signal.

According to another aspect, the invention provides a receiver unit comprising, for example, an associated board, a processor unit and an optical internet interface. Furthermore, the receiver unit may comprise a clock and trigger generator for providing all required clock signals to the antenna unit and the receiver unit. For example, the signal may be distributed through an optical waveguide or a coaxial cable.

For better integration, the correlation board (correlation board) can be implemented, for example, as a pure pci card, on which the functions required for the correlation calculations are combined. Thus, all existing venting and exhaust mechanisms of the disposer can be used well. The control link can also be applied directly to the processor unit by using a commercial optical internet board (10/100 Mbps).

For example, in the related board, all algorithms for determining the transmission time (synchronization) may be set as software modules, and ports may be set according to commercial computing platforms. Preferably, most of the algorithms are implemented as hardware modules on an FPGA platform (FPGA = field programmable gate array).

Further, the receiver unit may be configured to process signals from multiple transmitters (e.g., 150 transmitters). For example, if a signal is to be received from a particular transmitter, then according to the present invention, synchronization is performed for that particular transmitter using the inventive concepts described above.

As described above, part of the algorithm for determining the transmission time (synchronization) is implemented in hardware. For example, other required steps are performed by the processor. For example, the processor may be a PPC processor. However, the processor need not be a PPC; variations of conventional processors may also be used.

In an implementation there is one hardware module for each receiver communicating with the receiving unit of the invention. In contrast, the software may be implemented such that only one appears and estimates are made for all transmitters. The interface between the hardware and software modules may also be removed for easier implementation, if desired. For example, not in an FPGA (with rotation terms, phase relationships, etc.), but only in software, all partial correlation values are summed.

Next, the trigger and received signal windowing (part of the received signal) of the present invention is described.

The acquisition algorithm passes the pre-trigger signal a _ pre _ trigger, which accurately triggers a _ predist B _ sample, to the tracking algorithm before the tracking algorithm opens the next received signal window. The pre-trigger signal is activated after the a-burst is found. When triggering the window start, two situations occur depending on the implementation and the configuration of the transmitter. Fig. 3 shows the time relationship.

In the upper diagram of FIG. 3, t _A The starting time, t, of the A-burst in the received signal is described _F Is the correct start time for the window of the T-burst, T _P Is the time when the acquisition algorithm recognizes the a pulse train and can trigger the pre-trigger signal. The time period T = T is determined by the distance of an a-burst to the following T-burst _P -t _A The window starts at B _ sample before the relevant sequence of T bursts and is therefore specified by the configuration of the transmitter.

It is always assumed that at the input to the tracking algorithm, the pre-trigger signal occurs at B _ sample just before the desired window start. Depending on the implementation and the configuration of the transmitter, the situation in the upper and lower diagrams of fig. 3 occurs when the acquisition algorithm finds the a burst and has to start the window of the T burst.

If the distance t _F -t _P Greater than a _ predist (in units of B _ sample), the a algorithm will trigger the pre-trigger signal prematurely. In this case, the A algorithm must delay the pre-trigger signal to time T _P ', such that T _F -T _P ' = a _ predist, and passes the delayed trigger signal to the T-algorithm. In this case the FIFO buffer before the input of the tracking algorithm must never delay the input signal.

If the distance t _F -t _P Less than a _ predist, the acquisition algorithm triggers the pre-trigger signal too late. In this case, the FIFO buffer must delay the input signal by a _ predist- (t) _F -t _P ) So that after this the correct window start occurs at time t _F ' Here, t _F ’-t _P = a _ predist. Here, the acquisition algorithm finds the A-burst immediately (i.e., without delay) (i.e., at the time of the A-burst being found)Time t _P At) triggers the pre-trigger signal.

In fig. 4, as an example, four time signals are detected: the (undelayed) received signal basesig, the pre-trigger signal a _ pre _ trigger, the output signal of the FIFO and the position of the correlation sequence with which the T-burst is correlated in the receiver (in the partial correlation calculation).

The upper graph is based on the following assumptions: recisigdel (see fig. 4) is a positive value, meaning that the lower case in fig. 3 is about to occur. Again as described above, here the start of the A-burst, t, is plotted _A (measured in its first valid B _ sample, i.e. no transient phenomena of the filter), time of pre-trigger t _P Desired window start t 'in the (FIFO) delayed receive signal' _F 。

It should be noted that for implementation reasons, the T-burst starts with some padded zeros. However, the T-burst start is already the first B _ sample, i.e. the first padded zero.

The distance (1) between the start of two bursts (measured from the first significant B sample for a burst, without filter transients, on the other hand, from the first zero filled in the front end for T burst) is a _ dist _ B-a _ initdel (in B sample) and is therefore uniquely determined by the configuration of the transmitter. Here, at _ dist _ B is a value of at _ dist (in units of S _ cycle) converted into B _ cycle, and thus is not an integer.

Relatively, t _P -t _A Distance (2) of (a) is the only processing delay of the a algorithm, i.e. depending on the implementation. This value is a _ algoshift + a _ pipdel.

(3) Marking distance t' _F -t _P = a _ predist (in B _ sample).

At t' _F The windowing of the received signal begins and the correlation sequence is applied to the received signal "left aligned". Windowing is performed such that the expected received T burst is in phase with the correlation sequence at the calculated T _ nocorrvalsThe best overlap occurs in the middle of the off values. Thus, "advance (advance)" is t _ noncorrvals/2 correlation values, corresponding to advance (4), which is t _ noncorrvals = t _ noncorrvals/2B _ samples in the received signal window.

Finally, it must also be considered that the T-burst in the transmitter and the associated correlation sequence in the receiver are filled with zeros of different lengths. For the T burst X =0/1 of the considered transmitter, the length of the zeros padded at the beginning is the value tX _ startpadlen, and for the relevant correlation sequences in the receiver, the zero padding length is tX _ frontpaddlength. Thus, at tX _ frontpaddlength-tX _ startpadlen before the delayed T-burst, the correlation sequence has started, which is always the same as the sender for X =0 and 1. If there is a FIFO delay, the length of the padded zeros (5) must be considered. The received signal window must also be much longer than the T burst, since longer zeros are filled in the correlation sequence.

Furthermore, fig. 4 shows the delay (6) in the FIFO. As shown in fig. 5, the delay recisigdel (in units of B _ sample) may be calculated, dist _ Brnd being the rounding value of at _ dist _ B.

Fig. 6 shows a flow chart of the signal processing of the present invention:

the window start of the received signal can be triggered by the T algorithm itself or by the a algorithm.

To start a window, the a algorithm does not use a trigger signal, but a pre-trigger signal, which is activated a predetermined number a _ predist B _ samples before the window starts. By doing so to achieve the effect: even if there is a maximum frequency offset maxfreq offset ppm and a maximum length prediction in the T-algorithm (i.e., a prediction over T _ losttracthresh + 1T burst cycles, each T burst cycle being T _ burstperiod _ B _ sample), the pre-trigger signal of the A-algorithm always precedes the trigger signal of the T-algorithm, i.e., the trigger signal of the T-algorithm

A_predist＞＝(t_lostttrackthresh+1)*t_burstperiod_B* maxfreqoffsppm*10 ^-6 。

The activation of the pre-trigger signal and its function are explained in detail in the following examples.

The exit wait condition is triggered by the first trigger signal, i.e. by the T trigger signal or by the a pre-trigger signal;

once the a pre-trigger signal ends the wait condition, a _ predist B _ samples must be waited before the received signal windowing begins; in the case of a T trigger signal, window startup may be performed immediately;

in the main part of the sequence control, it is first determined which of the two T-bursts programmed into the small transmitter has been sent out (No. 1 or No. 2); correlation and all other calculations are performed for the parameters of the T-burst. To determine the transmitted T-burst, the flag T _ choice is used, which can be calculated in sequence control as described in [ RD6 ].

In order to correctly (from both possibilities) determine the T burst sent by the transmitter, it is necessary to carry a counter a _ multipl _ cntr as described in [ RD6] in the sequence control, however, it is sufficient to reset a _ multipl _ cntr to 0 in the receiver before the 1 st T burst of the acquisition cycle (and not before the A burst) as opposed to a small transmitter

In received signal windowing, t _ paddcorrseqlen + t _ nocorrvals-1B _ samples are copied into the buffer as the current received signal sample, where

O τ _ paddcorrseqlen is the length of the zero padded correlation sequence

Omicron t _ nocorrvals is the number of correlation values to be calculated

SNIR estimation in FPGA according to the following formula

SNIRset＝t_SINRcorrfact* maxsqcorr/abs(recenerg* t_corrsequenrg-maxsqcorr)

Here:

omicron SNITset is the estimated SNIR (in linear measurements)

Omicron Maxsqcorr is the maximum squared correlation magnitude measured in the partial correlation algorithm.

Omicabs () is a magnitude function for avoiding the numerical problem (if the denominator of the above term becomes negative).

Oenerrg is the energy measured in the partial correlation algorithm in the received signal over the length of the zero-padded correlation sequence.

Omicr _ corrseqenerg is the energy of the stored correlation sequence; this value depends on the transmission burst, i.e. the selection of the hardware module from two possible values when determining the transmitted transmission burst.

O τ _ snircorrerfect is a correction factor that takes into account the length of zeros padded in the correlation sequence; this value depends on the transmit burst, i.e. the selection of the hardware module from two possible values in determining the transmitted transmit burst.

When the estimated SNIRest is greater than the threshold T _ SNIRthresh, the T-burst is considered valid and/or found; the threshold value has to be chosen such that on the one hand a sufficient number of T-bursts are still considered valid, but on the other hand there is less probability of finding a T-burst erroneously at a location where there is no T-burst.

If a T-burst is found (SNIRest > = T _ SNIRthresh), the T-algorithm is synchronized again and starts predicting the next window start with its T-trigger signal; in addition, for more accurate analysis, the correlation values are passed to the software module "fine correlation".

The passing of the correlation values to the fine correlations represents a unidirectional interface between the FPGA hardware and software

Predicting the T trigger signal for the next window start according to the position of the correlation maximum found in the partial correlation algorithm by using the effective pulse train; adding T _ burst _ Brnd B _ samples to the position of the T burst period and decreasing T _ nopercorrvals B _ samples such that the window of width of T _ nocorrvals =2 × T _nopercorrvalscorrelation values is almost symmetric about the predicted next correlation maximum;

to confirm whether the T-algorithm is synchronized, a flag synclossflag is used, which is a value of 0 in case of synchronization, and 1 if the T-algorithm is not synchronized;

at synchronization, the T algorithm, using the variable nolosttracks, keeps track of how many last T bursts have been lost consecutively,

if no current T-burst is found (SNIRest < T _ SNIRthresh), verify if the T-algorithm is synchronized; if not, advancing to wait for the next A pre-trigger signal;

on the other hand, if the T-algorithm is still synchronized, a check is made to see if synchronization is still lost as the current T-burst is lost; if t _ lostttrackthresh has been continuously lost, i.e. nolosttracks > = t _ lostttrackthresh, synchronization is lost; in this case, with synclossflag = =1, the fine correlation software is ordered to inform the so-called ZRE of the loss of synchronization; so-called balers (ballers) in ZRE are therefore known: from this point in time until reacquisition, there is temporarily no TOA value from the receiver; furthermore, the T-algorithm deactivates its trigger signal so that the T-algorithm can only be triggered by a successful reacquisition of the a-algorithm;

if no more than T _ losttracking thresh T-bursts are continuously lost until the current T-burst (inclusive), a fine correlation for the calculated correlation value is activated even if the T-burst is not considered valid and may determine a very poor quality value; in addition, the T trigger for the next window start is predicted from the current window start: t _ burstperiods _ Brnd B _ samples (T burst periods) are simply added to this location.

Fig. 7 shows an example of the pre-trigger signal of the a algorithm of t _ lostratthresh = 2.

(1)t_burstperiod_B

(2)a_predist

(3) T trigger signal and correct window start in case of finding the last T burst

(4) Erroneous (predicted) T-trigger signal in case of no T-burst found

(5) Correct window start without finding a T-burst

(6) Pre-trigger signal of A algorithm derived from A pulse train

The figure shows how the worst case of the (predicted) trigger signal of the T-algorithm is generated from the correct window start: in (3), a T-burst is actually found last, where the T-trigger corresponds to the error window start. The next window start is now predicted over T _ burst _ B _ samples, in (4) the T trigger signal is triggered in each case. However, if there is a maximum frequency shift, then the correct window launch does not occur in (4), but rather occurs in (5), the farther away (4) the correct window launch is due to the extended T-burst period. Assume that at the time of triggering in (4), no T-burst is found. Predicting in (4') the last T trigger, i.e. T _ losttrakesh +1 (here three) T burst periods after the last T burst (3) found; since synchronization is lost after T _ lostttrackthresh + 1T bursts are lost in succession, no further prediction is made and the T trigger signal is deactivated from this point in time (now only the a algorithm can trigger). The correct window start associated with (4 ') is (5'); assume that the T burst is preceded by an a burst. Then the a pre-trigger signal must be generated in (6), i.e. before (4 '), so that the correct window start occurs in (5 ') instead of the T-trigger signal, which is incorrectly predicted in (4 '). (6) The distance between (a) and (5') is a _ predist > = (t _ lostthresh + 1) × t _ burst _ B × freqoffset ppm 10-6.

The processing delays that result from the execution of the acquisition algorithm and that have to be taken into account in the module FIFO are described below with the parameters a _ algoshift and a _ pipdel.

A _ aliasing indicates the difference (in B _ sample) between the start of the acquisition burst and the found correlation peak, resulting from the filter length used at different frequencies, the length of the acquisition burst and the parameter window _ total, while a _ pipdel indicates the delay in B _ sample resulting from the implementation of the receiver algorithm in hardware.

Fig. 8 shows the calculation of the difference.

Fig. 9 shows the basic construction of the inventive receiver for acquisition.

The receiver comprises a processing module 1101, the processing module 1101 having an input, and a plurality of outputs coupled to a filter 1103. The filter 1103 includes a plurality of outputs connected to a correlation block 1105. The correlation block 1105 includes an output connected to an oversampling device 1107. The oversampling device 1107 includes an output that is connected to another processing module 1109. The further processing module 1109 includes an output.

The signal received by the processing module 1101 is down-sampled and fs/4 mixing is performed, where fs represents the sampling frequency. The resulting signal is split into polyphase signals and the multiple phase signals are fed to a filter 1103 by a plurality of outputs, the filter 1103 may be a matched filter. For example, the filter 1103 includes a plurality of independent filters, each independent filter associated with a respective polyphase signal. From the filtered signal, a correlation is performed in a correlation block 1105 and the correlation result is then oversampled in an oversampling means 1107. The signal provided by the correlation block 1105 is fed into a further processing block 1109. The further processing module 1109 is arranged to calculate the position of the correlation maximum and to output a carrier signal.

The input 0-1 passes the received signal mixed to the complex baseband to layer 0-2 with the sampling clock B _ clock.

At outputs 0-7, for one of the 150 transmitters, the carrier signal obtained at layers 0-6 is passed with a sampling clock B _ clock _4 to a module that calculates partial correlation values with a tracking algorithm. In case a correlation maximum is detected, the value of the carrier signal is 1, otherwise the carrier signal is equal to 0. In order to be able to cut the part of the detected acquisition burst where the tracking burst is located after the acquisition burst, based on this carrier signal, it is necessary to delay the signal applied at the input 0-1, i.e. the "received signal", based on the total running time of the filters used, the time required for the acquisition algorithm, etc.

As described above, the receiver shown in fig. 9 is configured to perform acquisition (coarse synchronization) in the case where a plurality of receivers are to be synchronized. Thus, for example, 0-4, etc. has one output for each of 150 receivers. Thus, there are also 150 modules 0-5 and 0-6 and outputs 0-7.

According to another aspect, the present invention provides an apparatus for synchronizing a receiver with a transmitter configured to transmit a synchronization signal based on a coarse synchronization signal for coarse synchronization and a fine synchronization signal for fine synchronization.

The synchronization apparatus includes: sampling means for sampling a received version of the synchronisation signal to provide a received signal; signal processing means, as described above, configured to provide a position signal indicative of the position of the synchronisation signal in the received signal; and control means for controlling the sampling timing of the sampling means based on the position signal to synchronize the receiver and the transmitter.

Therefore, module synchronization can be achieved by controlling the sampling time of the sampling device.

In addition to the receiving structure, according to another aspect, the present invention provides an apparatus for generating a synchronization signal transmitted for synchronizing a receiver and a transmitter. The apparatus comprises: providing means for providing a coarse synchronization signal having a first bandwidth and a fine synchronization signal having a second bandwidth, the second bandwidth being smaller than the first bandwidth; and providing means for providing the synchronization signal using the coarse synchronization signal and the fine synchronization signal.

Here, the synchronization signal comprises a coarse synchronization signal and a fine synchronization signal and any desired combination of these two signals, the providing means being configured to provide the fine synchronization signal temporally after the coarse synchronization signal. This ensures that, when the synchronization signal is transmitted, a coarse synchronization signal for coarse synchronization in the receiver is transmitted before the fine synchronization signal, so that in the receiver, the coarse synchronization signal is initially detected to specify a part of the received signal where the fine synchronization signal is located, and synchronization is performed using the fine synchronization signal.

The providing means may be configured to combine the fine synchronization signal with the coarse synchronization signal such that the generated preamble (synchronization signal) comprises two parts. The providing means may be further configured to combine the plurality of fine synchronization signals and/or to combine copies of the plurality of fine synchronization signals in order to more accurately determine the possible locations of the fine synchronization signals in the receiver.

According to a further aspect, the providing means are configured for providing the fine synchronization signal temporally after the coarse synchronization signal such that a predetermined time interval exists between the provision of the coarse synchronization signal and the provision of the fine synchronization signal. For example, the time interval may be equal to a detection delay in the receiver and is introduced to enable detection of the fine synchronization signal after detection of the coarse synchronization signal in the receiver. Furthermore, the providing means may be configured to provide a plurality of fine synchronization signals temporally after the coarse synchronization signal. The fine synchronization signals may be directly consecutive or, as such, a time delay may be provided between the fine synchronization signals, wherein the distance between consecutive fine synchronization signals is of a magnitude that determines the delay with which the position of the fine synchronization signal in the portion of the received signal is determined.

The providing means are further configured to provide the plurality of fine synchronization signals by combining the copies of the fine synchronization signals in time, e.g. as a sequence, such that there is no delay between the fine synchronization signals, as described above.

According to another aspect, the providing means comprises a burst shaping filter for filtering the coarse synchronization signal and/or the fine synchronization signal. The burst shaping filter may also be configured to filter the entire synchronization signal. For example, the burst shaping filter may be a cosine roll-off filter with a roll-off factor of 1. In this case, the filter characteristic has a pure cosine shape in the frequency range.

According to another aspect of the present invention, the burst shaping filter has a variable characteristic, and thus the fine synchronization signal and the coarse synchronization signal may be differently filtered.

According to another aspect, the providing means is arranged for generating a coarse synchronization signal from a data sequence and a fine synchronization signal from a further data sequence, the data sequence having a smaller bandwidth than the further data sequence.

According to an embodiment, the providing means comprises a memory in which said data sequence and said further data sequence are stored and from which they can be retrieved. Here, for example, the coarse synchronization signal may be identical to the data sequence, and the fine synchronization signal may be identical to the other data sequence.

The providing means may further comprise a generator configured to generate said data sequence and/or said further data sequence. For example, the generator may be configured to generate the data sequence from a Galois field (Galois) having four elements. Furthermore, the generator may be configured to generate the further data sequence from a galois field having four elements or more than four elements. However, galois fields having two elements may also be used.

For generating the data sequence and/or for generating the further data sequence, the generator may comprise a shift register, the generator being configured to set an initial occupancy of the shift register, so that depending on the respective initial occupancy a plurality of data sequences and a plurality of further data sequences may be implemented.

The providing means may further comprise correlating means for associating a complex value with each element of the data sequence to obtain a sequence of complex values and/or associating a complex value with each element of another data sequence to obtain another sequence of complex values. In other words, the correlation means are configured to map the data sequence to a complex-valued sequence and to map the further data sequence to a further complex-valued sequence, the data sequence and the complex-valued data sequence having the same number of coefficients and the further data sequence and the further complex-valued sequence having the same number of coefficients. The number of coefficients of a complex-valued sequence may be different from the number of coefficients of another complex-valued sequence. However, the two complex-valued sequences may also comprise the same number of coefficients.

The coarse synchronization signal and/or the fine synchronization signal may be a band pass signal or a baseband signal. If the coarse synchronization signal is a band pass channel, the providing means further comprises an up-converter for generating the coarse synchronization signal by up-converting the complex sequence. The up-converter may be further configured to generate a fine synchronization signal by up-converting another complex-valued sequence. Since the bandwidth of the coarse synchronization signal is narrower than that of the fine synchronization signal, the coarse synchronization signal occupies a band-pass range having a center frequency different from another center frequency of another frequency range occupied by the fine synchronization signal. However, according to another aspect, the center frequency and the further center frequency may be the same.

According to another aspect, the providing means may be configured to provide a further fine synchronization signal having a bandwidth larger than the bandwidth of the coarse synchronization signal. The further fine synchronization signal may be different from the fine synchronization signal and may be orthogonal to the fine synchronization signal. The means are thus provided for processing the further fine synchronization signal as said fine synchronization signal, as described above.

According to another aspect, the providing means may be configured to encode the information with a phase variation between successive values of the fine synchronization signal. For example, the providing means are configured for encoding the information with a 180 ° phase jump between a value of the fine synchronization signal and another value of the fine synchronization signal. For example, one bit is transmitted in addition to the synchronization information by phase hopping. However, according to another aspect, the fine synchronization signal may include a plurality of phase jumps, thus encoding the information sequence.

According to another aspect, the providing means may be configured to encode the information with a phase change, e.g. a 180 ° phase jump, between consecutive values of the coarse synchronization signal. In this way, the receiver is able to detect the additional information by phase change detection.

The apparatus for generating a synchronization signal of the present invention may further include a transmitter for transmitting out the synchronization signal. The transmitter may be, for example, a radio transmitter.

According to another aspect, the present invention provides a transmission apparatus having: a device for generating a synchronization signal as described above; control means for controlling the supply means so as to generate a predetermined time sequence of the coarse synchronizing signal and the fine synchronizing signal; and a transmitter for transmitting a synchronization signal comprising a time sequence of a coarse synchronization signal and a fine synchronization signal.

According to the invention, the transmitter is configured to transmit a narrowband acquisition burst with which low overhead synchronization (acquisition) with the signal of the transmitter can be performed. Using the A-burst, the position of the subsequent T-burst is predicted. The transmitter sends out a wideband tracking burst. This allows accurate measurement of the arrival time of the tracking burst in the receiver. The transmitter also transmits data that has been entered into a T-burst (entered data transmission) at a low data rate. The phase change described above may be used to enter data.

In future toy systems, for example, the transmitter may also be configured to transmit data using a burst transmission system, which data may be received with very low overhead (burst-wise data transmission).

Acquisition bursts (coarse synchronization signals) are used to synchronize the receiver with each transmitter in a multi-transmitter, multi-receiver scenario, with very low overhead. An approximate estimate of the transmission time of the tracking burst used to provide the location for determination is obtained. In the case of lost tracking, the acquisition burst is used for both initial synchronization (acquisition) and resynchronization (reacquisition).

If a receiver is to be synchronized with multiple transmitters, the transmitters must be distinguishable. The transmitters are each configured to transmit out the inventive synchronization signal, and the respective transmitters preferably use different sequences of a certain length to be able to distinguish. To implement the receiver at low cost, the bandwidth of these sequences is much narrower than the bandwidth of the tracking bursts (fine synchronization signals). However, to be able to achieve a good enough distinction, each transmitter may use 9 different "subcarriers" of the used frequency band.

The allocation of the relevant parameters (sequence, frequency) can be implemented, for example, by a central database which ensures that no two transmitters use the same sequence at the same frequency. In addition, other boundary conditions may occur in the sequence assignment, which are eliminated using external programming.

For example, the transmission sequence (data sequence and/or another data sequence) can be generated with a generator using a software program, with which the digital signal processing and/or the signal transmission is simulated. For this purpose, for example, a transmission sequence generated from GF4 (galois field having four elements) is used. For example, a polynomial for generating a sequence can be found in "4-Phase-Sequences with near-optimal Correlation Properties" by Serdar Boztas, roger Hammons and Vijay Kumar, IEEE Transactions on Information Theory, vol.38, no.3, may 1992, pages 1101, and the like. For example, a given polynomial may be employed and its order may be changed, e.g., 32113111 may be generated in software from a given [11131123 ].

The generator polynomial is defined as [32113111]. By changing the initial occupancy of the shift register, a plurality of sequences (data sequences) are obtained. It should be noted that no internal register values are generated from the already generated sequences, since the generated sequences are no longer irrelevant in this case.

For example, the symbol rate is designated as B _ clock _48,i.e. f _symb =1/48 × b lock. For example, the length of a transmission sequence of an arbitrary rate is 511 symbols.

As described above, the acquisition burst may be formed over a frequency spectrum. For this purpose, a sequence of elements of the set of {0,1,2,3} is mapped, filtered and transmitted at a specific carrier frequency.

In mapping (correlation), a correlation symbol is generated from an element using a mapping. As an example, fig. 10 shows a specification according to which the mapping sequence is based.

Fig. 11 shows a constellation diagram, in which the specification of points is shown.

The burst shaping is achieved by a cosine filter with a roll-off factor α =1.0 rising at the square root. The related formula of the transfer function is as follows:

h (f) =1 for | f | < f _N (1-α)

For f _N (1-α)＜|f|＜f _N (1+α)

H (f) =0 for | f | > f _N (1+α)

Fig. 12 shows the transfer function of the filter.

For example, the complex signal generated and present at the minimum sampling rate of 2 × b clock _48is now distributed to 9 different frequency bands. The distribution may be performed by the upconverter described above.

Fig. 13a shows the distribution of a complex signal to 9 different frequency bands. The next carrier frequencies are selected and numbered according to the table shown in fig. 13 b. This information is related to the carrier frequency of the 2445MHz T-burst.

The T-burst may be programmed in the transmitter. After being sent out, the T-burst exhibits the following characteristics, for example: its bandwidth does not exceed the value t _ burst tbw, its duration is approximately the adjustable value t _ burst _ cycles (plus additional time extensions in the conversion from the programmed B _ sample to the physical signal (e.g. interpolation, dispersive analog circuits, etc.)), the maximum duration is t _ burst _ xlen B _ cycles (plus the above extensions), the maximum duration is predetermined by the implementation of the transmitter, the SNIR after the transmitter output has the value of the transmitter SNIR or better.

According to the invention, two T-pulse trains can be distinguished from each other by: each T-burst exists as it is generated before programming, i.e. without zero padding, and is programmed (in the transmitter) according to the specific situation, performing length adjustment with zero padding.

In the transmitter, two programmed T-bursts are transmitted according to a programmable selection scheme. These two T-bursts are referred to as T-burst 0 and T-burst 1, respectively. Each of these two T-bursts is present as a complex-valued B _ sample, i.e. the T-burst signal to be transmitted is sampled in the I and Q components at the sampling frequency B _ clock, respectively.

T-bursts are generated so that they are optimally controlled with the 6-quantization used. I.e. the signal actually assumes a maximum representable value of + 31.

The length of the first T-burst 0 and 1 programmed in the transmitter can be adjusted jointly, i.e. they are always the same and they are equal to T _ sendbacklen (in B _ sample). The two T-bursts may exhibit slightly different lengths T _ burst (for T- bursts 0 and 1, respectively), differing by only a few B _ samples, before programming to the transmitter. With zero padding (see below), different lengths of the pulse train are adjusted. The maximum length of a T-burst is T _ burst max len (in B _ sample).

The granularity at which the length of the programmed T-burst can be changed is 8B samples (for implementation reasons), i.e. the burst length to be zero-padded (zero-padded) is a multiple of 8 for the transmitter. The shorter one is filled with more zeros than the longer one of the two T-bursts, where a maximum of 7 zeros need to be filled. If tX _ burst is the length of the original T-pulse string X (no zero padding), then the following applies:

T_sendburstlen＝ceil(max(t0_burstlen，t1_burstlen)/8)*8

in order to center the original T-burst X (without additional zeros) almost at the center of the programmed zero-padded burst tx _ sendbusst (required to enter the channel bits to be transmitted),

tX_startpadlen＝ceil((t_sendburstlen-tX_burstlen)/2)

at the beginning of the T-burst, zeros are filled and at its end the remaining zeros, i.e. zeros are filled

floor ((t _ sendbistlen-tX _ burst)/2) zeros. Thus, the same number of zeros is padded at the end of the burst as at the beginning thereof, and/or most likely an additional zero is padded at the end thereof compared to the beginning thereof.

For example, the acquisition bursts and tracking bursts may be transmitted in time division multiplexing.

In the multiplexing of a and T bursts, one a burst is sent exactly in one acquisition cycle, followed by at _ multipll _ len T bursts, the value of at _ multipll _ len being adjustable. The scheme continues periodically. An adjustable distance is maintained within a cycle between an a-burst and a first T-burst, between respective T-bursts, and between a last T-burst of the cycle and an a-burst of a next cycle.

There is a counter at _ multipli _ cntr in the transmitter, which is always reset to 0 before the start of an a-burst and incremented by 1 after each T-burst has been completely transmitted. Therefore, the maximum value of at _ multipl _ cntr is at _ multipl _ len.

For time division multiplexing, a burst of a pulses occurring in units of B _ sample is considered.

In the following, when referring to "related a-bursts" (always in B _ sample) only the part described below (always in B _ sample) of the "generated a-burst" generated from the a-burst using transmit burst shaping, upsampling and upconversion (either using Matlab or directly in the transmitter itself) is indicated, as is present at the multiplexed input.

Fig. 14 shows a portion of the generated a-burst referred to as "related a-burst". The upper half of the image shows the "root a-burst" in units of B _ sample _48 (so the distance between two samples (1) is B _ cycle _ 48). Its entire length (6) is a _ burst (in units of B _ cycles _ 48).

The delay and duration of the resulting a-burst in B _ sample (with transmit burst shaping, oversampling and upconversion), the delay and duration of the stored a-burst in B _ samples _48 and the related a-bursts are shown.

The lower half shows the resulting a-burst in units of B _ sample (thus the distance (2) between two samples is B _ cycle).

At time t0, the first B _ sample _48 input signal of the stored a-burst is generated (transmit burst shaping, upsampling, and upconversion). Depending on the implementation, this first B _ sample _48 of the stored a-burst does not appear in the generated a-burst until time t2, i.e. after delay (3). The delay includes two parts: buffering in signal generation for pipelining and the like occurs from time t0 when the first B _ sample _48 of the stored A burst is passed into signal generation to time t7 when the first valid B _ sample of the A burst generated by the signal generation output is generated (4). Furthermore, all the filters used (transmit burst shaping and upsampling lowpass) have a group running time (group runtime) which cumulatively forms the filter running time (5). If the generated T-burst is generated offline using Malab, then delay (4) =0 unless the transmitter introduces an additional pipeline. However, since the delay (i.e., group running time) caused by the setting of the filter is always present, it is considered.

If the last B _ sample _48 input signal of the stored a-burst is generated at time t3, it will not appear at the output of the signal generation as B _ sample either before time t4, after delay (3). At time t5, the last valid B _ sample of the a-burst generated by the signal generation output. After duration (7), i.e. after t5-t4, all filters attenuate. Since the usual filters have a symmetrical burst response, the duration (7) is equal to the accumulated group running time (5).

The cumulative burst broadening (i.e. the sum of the fall and decay times, (5) + (7)) due to the wave dispersion filter must not exceed the value a _ maximum width B _ sample for correct operation of the whole system, which is ensured especially for the receiver.

The overall length of the resulting a-burst (in terms of the effective B _ sample) is therefore (8), i.e. the sum of (5), (6) and (7). Since (5) and (7) depend on the implementation, it always regards (6) (i.e. a _ burst, denoted B _ samples _ 48) as the length of the associated a-burst. The other expressions are: in units of B _ sample, a _ burst _ B is the distance between the first and last B _ samples, which belong to the first and last B _ samples _48 of the stored a burst, respectively.

Thus, the length of the a-burst considered below does not contain signal broadening due to the burst response of the dispersion filter in the signal generation (transmit burst shaping and up-sampling low-pass) and does not contain signal broadening introduced by possible other implementations.

The scheme shown in the lower part of fig. 14 shows multiplexing, burst duration and distance. The illustrated scheme reflects burst multiplexing in the transmitter at precisely the positions specified below; all time scales shown in the lower part apply to the multiplexer, which switches between a-bursts and T-bursts in the transmitter. At this position, both a and T bursts exist in units of B _ sample.

The term T-burst refers to two programmed T-bursts to be selected as described above.

Fig. 15 shows the duration and distance in burst multiplexing.

The relevant a-burst exhibits a length a _ burst (2) (in units of B _ samples _ 48) (see the explanation above regarding a-bursts and their length), which must be shorter than the maximum length a _ burst max (1) (in units of B _ cycles _ 48) of the relevant a-burst.

The T-burst exhibits a length T _ burst (4) (in B _ cycle) which must be shorter than the maximum T-burst length T _ burst _ tmax len (3) (in B _ cycle).

The pause between the end of the burst of a and the start of the burst of T should be kept at a fixed value even if different lengths of (a and/or T) bursts are used in the system. If the relevant a-burst is made shorter or longer, only its start (front stop) should be moved accordingly. If the T-burst is to be made shorter or longer, only its tail should be moved accordingly.

The distance from the start of the relevant a-burst to the start of the following T-burst can be adjusted using two parameters. The distance at _ dist (5) from the generation of a burst of trigger a to the start of the following T burst can be adjusted in units of S-cycles. Furthermore, the initial delay a _ initdel (9) from triggering the start of the relevant a-burst (according to the above definition) can be adjusted in units of B-cycles. Thus, the actual distance from the start of the relevant A burst to the start of the T burst is at _ dist [ s _ cycle ] -a _ initdel [ B _ cycle ].

For implementation reasons, the distance at _ dist is in units of S _ cycle.

For implementation reasons, the distance T _ burst (6) between the start of two consecutive T bursts of an acquisition cycle can also be adjusted in units S _ cycle.

The distance ta _ dist (7) between the start of the last T burst of an acquisition cycle and the trigger before the generation of the a burst of the next acquisition cycle can also be adjusted in units of S-cycles.

In the transmitter, the actual distance may vary slightly around its nominal value for implementation reasons.

These three burst distances are not freely adjustable, but must satisfy the following requirements:

ta_dist+at_dist＝t_burstperiod

the distance a _ burst period (8) between the start of the a-bursts of two consecutive acquisition periods is not adjustable in the transmitter but by other adjustable parameters:

a_burstperiod[S_cycle]＝at_multipl_len*t_burstperiod[S_cycle]

for example, the T-burst period is expressed as a multiple of 100B _ samples/reference frequency. These multiples are relatively prime to avoid adverse re-overlapping conditions. For example, if the burst period is about 500 milliseconds, the following period (in units of B _ sample) is appropriate: 100B _ sample × 463 467 479 487 491 499 503 509 521 523 (the first prime number is 541). The minimum period is therefore 4545 ms (i.e. 2200/s). The maximum period is 5134 milliseconds (i.e., 1948/s).

Orthogonal transmit bursts may also be designed according to a pseudo-random principle, which may be based on a prime number algorithm. According to the invention, the transmitters may be assigned a fixed burst repetition rate, which, however, differs slightly between different transmitters (non-synchronized pseudo-random pattern). Therefore, there is a method according to which the burst distance of the transmitter is randomly selected to avoid overlapping of fixed patterns of bursts. For example, one transmitter transmits 2000 bursts per second, while another transmitter transmits 2001 bursts per second.

One is always selected from two T bursts stored in the transmitter to transmit. The selection is made with a bit t _ choice, which is obtained from a counter at _ multipll _ cntr with a programmable mask t _ choice _ mask according to the following specification.

T_choice＝(t_choice_mask[0]AND at_multipl_cntr[0])OR

(t_choice_mask[1]AND at_multipl_cntr[1])

<xnotran> , [0] ( LSB ), [1] , , AND / OR AND / OR. </xnotran>

If T _ choice represents a value of 0, T burst 0 is transmitted, and when T _ choice = =1, T burst1 is transmitted.

For T _ choice _ mask = =00 (LSB on right hand side), a T burst of 0 is always transmitted;

for T _ choice _ mask = =01, T burst 0 and T burst1 are transmitted in sequence after each T burst (starting with T burst 0 after a burst)

For T _ choice _ mask = =10, T burst 0 and T burst1 are transmitted after every other T burst (starting with T burst 0 after a burst), i.e. the transmission: t burst 0,T burst 0,T burst1,T burst1,T burst 0,T burst 0,T burst 1.

T _ choice _ mask = =11 is not a useful choice, and cannot be set.

There may be two transmit antennas in the transmitter. Both transmit antennas may be used simultaneously or only one may be selected. If both are used, the programmable bit two _ ants is set to a value of 1. If two _ ants = =0, only one of the two transmit antennas is selected to transmit in any case. The selection is made with a bit ant _ choice obtained from the counter at _ multipl _ cntr using a programmable mask ant _ choice _ mask according to the following specification:

Ant_choice＝(ant_choice_mask[0]AND at_multipl_cntr[0])OR

(ant_choice_mask[1]AND at_multipl_cntr[1])。

the variable [0] represents the zeroth bit (i.e., LSB) of the variable, the variable [1] represents the first bit, AND further, AND/OR means AND/OR bit by bit.

If ant _ choice has a value of 0, it is transmitted using antenna 0, and if ant _ choice = =1, it is transmitted using antenna 1.

However, this only applies to two _ ants = =0. When two _ ants = =1, each burst is sent out on two antennas. For two _ ants = =0, with the following scheme:

for ant _ choice _ mask = =0 (LSB on right-hand side), always transmit using antenna 0,

for ant _ choice _ mask = =01, using antenna 0 and antenna 1 after each T burst in turn (for an a burst and subsequent T burst, starting with antenna 0)

For ant _ choice _ mask = =10, antenna 0 and antenna 1 are used after every other T burst (for an a burst and the following T burst, starting with antenna 0), i.e. transmission is performed on the antennas one after the other in the following way: antenna 0, antenna 1, antenna 0, antenna 1.

Ant _ choice _ mask = =11 is not a useful choice, and cannot be set

As mentioned above, the T-burst can also be used to initiate the transmission of data (of other information) keyed to the T-burst on a bit-by-bit basis.

The keyed data transmission is performed in a frame-based manner. A frame is a selection of useful data that is jointly transmitted within a time interval. For this transmission, frame _ len T bursts are required, the value frame _ len being adjustable. Subsequently, the next frame is transmitted in one time interval. The choice of useful data to be transmitted is the same for different frames, only the content of which can be changed. Each frame is independent and can be received and estimated, i.e. synchronized and decoded, independently.

In the transmitter, there is a bit oddframeflag, which is always switched (from 0 to 1 or from 1 to 0) immediately before the next frame is generated, i.e. it has a value of 0 at even frame numbers and 1 at odd frame numbers.

Compilation of the data corresponds to reading out the register selected for transmission, the selection being programmable. The compilation, i.e. the reading out of the selected register, must be done in as short a time period as possible. The data of the selected register then exists as a single long bit vector databitvec. The entire amount of data, i.e., the length of databtvec (in bits) is referred to as the nondatabits.

In order to add the cyclic redundancy check bits, the invention also provides a symmetric encoder. Fig. 16 shows a block diagram of a symmetric encoder for a CRC-12 code.

In order for the receiver to be able to perform an integrity check of the received data and to be able to detect any residual error after channel decoding, 12 parity bits of the CRC-12 code are calculated for the data in the form of databitvec and attached after the end of the databitvec. The bit vector thus generated is referred to as a checkdbitvec. In order to be able to use such a CRC-12 code, the nondatabits must be 2035 or less.

The CRC-12 code is characterized in that it generates a polynomial g (X) = X ¹² +X ¹¹ +X ³ +X ² +X+1。

The encoder contains 12 interconnected registers, each of 1 bit. All additions are made in binary form, i.e., 1+1=0 (EXOR). Furthermore, two switches are included, which are connected to each other. At the beginning, both switches are in the lower position. Before starting, all 12 bit registers are initialized to a value of 0. The bits contained in the databitvec are then applied one after the other at the input, i.e. one bit per module. At the output, the resulting bits are simultaneously read out of the checkdbitvec. First, databitvec [0] is applied at the input (index starts with 0). The checkedbitvec [0] is read out at the output. After the nodal-1 clock, the last input bit, databitevec [ nodal-1 ], is applied at the input and the checkedbitvec [ nodal-1 ] is read out at the output. For the next clock, both switches are switched to the upper position. From this point on, the line below the register constantly has a bit value of 0. The output is still read out within 12 clocks including this clock and a checkdbitvec [ nodatabits ] to a checkbitvec [ nodatabits +11] is generated. Then, the generation of CRC is completed.

In summary, the resulting bit vector, checkdbitvec, has a length nocheckbits = bodatabits +12, the first bodatabits being the same bits as the vector databitvec.

According to the present invention, as a measure for protecting against transmission errors, channel coding (forward error correction code) is also performed. FIG. 17 shows a block diagram of a 1/2 rate time-limited recursive systematic convolutional encoder.

As a measure for protecting against transmission errors, a forward error correction code is used to perform channel coding of the vector checkdbitvec. Specifically, this is a rate 1/2 time-limited recursive systematic convolutional code, from the industry standard convolutional code [ RD 2]]Characterized in that the generator polynomial is (1, (x) ⁶ +x ³ +x ² +x+1)/(x ⁶ +x ³ +x ² + x + 1)). By encoding the vector checkbitvec, a vector codebitvec including code bits is generated.

The encoder contains 6 interconnected registers of 1 bit per bit. All additions are made in binary, i.e. 1+1=0 (EXOR). At the beginning, the included switch is in the upper position. Prior to this state, all 6-bit registers are initialized to a value of 0. The bits contained in chechedbitvec are then applied to the input exactly one after the other, in such a way that a different bit is applied every 2 clocks. The contents of the 6 registers are also only changed when a different bit is applied to the input every 2 clocks. The two outputs are read out one after the other in parallel, the upper output being read out first and the lower output being read out at the next clock. Initially, checkdbitvec [0] is applied at the input (indexing starts with 0), and, initially, codebitvec [0] (at the output above) is read out, and in the next clock, codebitvec [1] (at the output below). After a total of two clocks, the checkedbitvec [1] is applied at the input and in the subsequent clock, codebitvec [2] and codebitvec [3] are read out at the output. After a total of 2 clocks (nochedbits-1), chedbitvec [ nochedbits-1 ] is applied at the input and codebitvec [2 x nochedbits-2 ] and codebitvec [2 x nochedbits-1 ] are read out at the output. After a total of 2 nocheckeldbits clocks, the switch is switched to the lower position. Furthermore, the contents of the register may only change once every 2 clocks. Including the clock, the outputs are optionally read out over a total of 12 clocks and produce codebitvec [2 × nocheckbits ] through codebitvec [2 × nocheckbits +11]. This completes channel coding. With the shown implementation of the encoder, finally, all registers contain again a 0-bit value.

The total length of the generated bit vector codebivtvec is nocodebits =2 nocheckeldbits +12, and the bits with indices 0, 2, 4,. ·,2 · (nocheckeldbitcs-1) are the same as the vector checkdbitvec.

As a measure for protecting against deletion of a plurality of consecutive code bits, the order of code bits contained in codebivec is changed in a channel-interleaved manner. Here, a bit vector ilvedbitvec to be output is generated. The interleaver used is a modulus interleaver. The interleaver performs the following changes to the bit order:

ilvedbitvec [ i ] = codebitvec [ (i. Ilvshift) mod nocodebits ] wherein i =0

ilvshift is a tunable value, and the index of the bit vector starts at 0.

Finally, the length of ilvedbitvec is nocodebits (in bits).

The preamble for frame synchronization is placed before the ilvedbitvec starts:

if oddframeflag = =0, the Barker sequence 11100010010 is used as the pre-synchronization code

If oddframeflag = =1, then the inverse Barker sequence 00011101101 is used as the preamble

The preceding preamble followed by ilvedbitvec yields a vector of channel bits, referred to as channelbitvec. Its length is frame _ len = nocodebits +1.

All the fram _ len bits of the vector channlbitvec are sent in T bursts independently, i.e. frame _ len T bursts are needed to send the entire vector channlbitvec. The mapping of each channel bit and the shaping of the transmission burst are performed in one single step, i.e. in a step of keying in the T-burst using the differential binary phase shift keying direct sequence spread spectrum (DBPSK-DSSS) method.

Each T-burst is generated from half of the bursts halfburst1 and halfburst2 in the form of B _ sample (having I and Q components, respectively).

Fig. 18 shows the generation of a T-burst.

To generate the T-burst, the present invention provides apparatus comprising means for generating the T-burst with the check channel bits entered.

Half of the bursts halfburst1 and halfburst2 each have a length halfburstlen (1) (in units of B _ sample). To generate a T-burst, halfburst2 is delayed in time, weighted by +1 and-1, and added to halfburst 1. The time delay of halfburst2 relative to halfburst2 is shift12 (in units of B _ sample). halfburstlen and shift12 have different values for each programmable T-burst.

The transition region where halfburst1 and halfburst2 (the latter delayed in time by shift 12) overlap has a length of halfburstlen-shift 12B _ samples, always obeying the constraint halgburstlen-shift12= < overlapmaxlen (in units of B _ samples) in the design of a half burst.

The weighted addition of halfburst2 and halfburst1 to perform time delay is performed according to the following scheme:

if the channel bit to be currently transmitted is 0 value, add to + halfburst 2;

if the channel bit to be currently transmitted is 1 value, add to-halfburst 2.

Therefore, generation of a T burst having a length T _ burst = halfburstlen + shift12 for the current channel bit is completed and the burst can be transmitted.

Depending on the situation, the inventive method can be implemented in hardware or in software. The implementation can be implemented on a digital storage medium, in particular on a disc or a CD having electronically readable control signals, which cooperate with a programmable computer system such that the corresponding method is performed. Generally, therefore, the present invention also includes a computer program product with a program code stored on a machine-readable carrier, which when run on a computer performs the inventive methods. In other words, the invention can thus be implemented as a computer program having a program code for performing the method when the computer program runs on a computer.

Claims (45)

1. A synchronization apparatus for determining a position of a synchronization signal in a received signal, the synchronization signal being based on a coarse synchronization signal and a fine synchronization signal, the synchronization apparatus comprising:

signal processing means configured to determine a part of the received signal, in which the fine synchronization signal is located, based on the coarse synchronization signal and to determine a position of the synchronization signal in the received signal in the part of the received signal for synchronization based on the fine synchronization signal,

wherein the signal processing means are arranged for detecting a fine synchronization signal in said part of the received signal, an

Wherein the bandwidth of the coarse synchronization signal is smaller than the bandwidth of the fine synchronization signal, and the signal processing means is configured to perform the detection of the coarse synchronization signal in the received signal at a first sampling rate and to perform the detection of the fine synchronization signal at a second sampling rate, the second sampling rate being higher than the first sampling rate.

2. A synchronization device as claimed in claim 1, wherein the signal processing means are arranged for detecting a coarse synchronization signal in the received signal in order to determine said part of the received signal.

3. A synchronization device as claimed in claim 1 or 2, wherein the synchronization signal comprises a coarse synchronization signal and a fine synchronization signal, the signal processing means being arranged to detect the coarse synchronization signal in order to determine the start of the part of the received signal at which the fine synchronization signal is located.

4. A synchronization device as claimed in one of claims 1 to 3, wherein the signal processing means are arranged for performing a correlation between the received signal and a signal related to the coarse synchronization signal in order to detect the coarse synchronization signal in the received signal, and for performing a correlation between the received signal and a signal related to the fine synchronization signal in order to detect the position of the fine synchronization signal in said part of the received signal.

5. Synchronization device according to one of claims 1 to 4, wherein the signal processing means comprise first detection means (101) and second detection means (103), the first detection means (101) being arranged for detecting a coarse synchronization signal in the received signal and the second detection means (103) being arranged for detecting a position of a fine synchronization signal in said part of the received signal.

6. A synchronization device as defined in claim 5, wherein the first detection means are configured to detect the coarse synchronization signal at a first sampling rate and the second detection means are configured to detect the position of the fine synchronization signal at a second sampling rate, the first sampling rate being lower than the second sampling rate.

7. A synchronization device as claimed in claim 5 or 6, wherein the coarse synchronization signal occupies a predetermined frequency range, the first detection means comprising a filter for filtering the received signal to filter out received signal components occupying the predetermined frequency range.

8. The synchronization apparatus of claim 7, wherein the filter is a band pass filter.

9. A synchronisation device as claimed in claim 7 or 8, wherein the first detection means comprises a down-converter for down-converting the received signal component.

10. Synchronization device according to one of claims 7 to 9, wherein the first detection means comprise a detector configured for detecting a coarse synchronization signal in the received signal component.

11. A synchronization device as claimed in claim 10, wherein the detector is configured to perform a correlation between the received signal component and a signal related to a coarse synchronization signal.

12. The synchronization apparatus of claim 11, wherein the detector is configured to output a detection signal indicating detection of a coarse synchronization signal when the correlation value exceeds a detection threshold.

13. Synchronization device according to one of claims 5 to 12, wherein the first detection means are configured to provide a detection signal indicating a detection time at which the coarse synchronization signal is detected, the detection time indicating a start of the part of the received signal at which the fine synchronization signal is located, and the second detection means are configured to receive the detection signal and to detect a position of the fine synchronization signal in the part of the received signal in response to the detection signal.

14. A synchronization device as claimed in one of claims 5 to 13, wherein the second detection means comprise delay means for delaying the received signal, said delay means being configured to compensate for the detection delay of the first detection means.

15. A synchronisation device as claimed in any one of claims 5 to 14, wherein the first synchronisation signal comprises a predetermined bandwidth and the second detection means comprises a filter for limiting the received signal to within the predetermined bandwidth.

16. Synchronization device according to one of the claims 5 to 15, wherein the second detection means comprise a detector for detecting the position of the fine synchronization signal in said part of the received signal.

17. A synchronization device as defined in claim 16, wherein the detector is configured to determine a correlation of the received signal with a signal related to the fine synchronization signal in order to detect the position of the fine synchronization signal.

18. The synchronization apparatus of claim 17, wherein the detector is configured to determine a first partial correlation between a first subset of received signal values in the portion of the received signal and a first subset of signal values related to the fine synchronization signal, and to determine a second partial correlation between a second subset of received signal values in the portion of the received signal and a second subset of signal values related to the fine synchronization signal, such that the first partial correlation and the second partial correlation overlap are used to determine the correlation.

19. The synchronization device of claim 18, wherein the detector is configured to detect a phase relationship between respective values of the first partial correlation and the second partial correlation, the phase relationship including a frequency shift between the transmitter and the receiver, and the detector is configured to cancel the phase relationship by weighting the value of the first partial correlation or the value of the second partial correlation so as to reduce an effect of the frequency shift on the correlation values.

20. A synchronization device as claimed in one of claims 17 to 19, wherein said detector comprises an interpolator for interpolating between correlation values to obtain fine correlations and for detecting the position of the fine synchronization signal with a higher accuracy from the fine correlations.

21. Synchronization device according to one of the claims 5 to 20, wherein the fine synchronization signal comprises information encoded by a phase relation between consecutive values of the fine synchronization signal, and the second detection means (103) are configured for detecting the information by detecting the phase relation.

22. Synchronization device according to one of the claims 17 to 21, wherein the second detection means (103) are configured for deriving the quality value of the reception quality from the correlation.

23. Synchronization device according to one of the claims 1 to 22, wherein the second detection means (103) are configured for outputting a position signal indicative of the position of the fine synchronization signal in said part of the received signal, in order to indicate the position of the synchronization signal in the received signal.

24. An apparatus for synchronizing a receiver and a transmitter, the transmitter configured to transmit a synchronization signal based on a coarse synchronization signal for coarse synchronization and based on a fine synchronization signal for fine synchronization, the apparatus comprising:

sampling means for sampling a received version of the synchronisation signal to provide a received signal;

a synchronization device as claimed in one of claims 1 to 22, the synchronization device being configured to provide a position signal indicating a position of the synchronization signal in the received signal; and

control means for controlling the sampling time of the sampling means based on the position signal so as to synchronize the receiver and the transmitter.

25. An apparatus for generating a synchronization signal, which is transmitted to synchronize a receiver and a transmitter, the apparatus comprising:

providing means for providing a coarse synchronization signal having a first bandwidth and a fine synchronization signal having a second bandwidth, the first bandwidth being smaller than the second bandwidth; and

providing means for providing a synchronization signal using the coarse synchronization signal and the fine synchronization signal.

26. The apparatus of claim 25, wherein the synchronization signal comprises a coarse synchronization signal and a fine synchronization signal, the providing means being configured to provide the fine synchronization signal temporally after the coarse synchronization signal.

27. Apparatus according to claim 25 or 26, wherein the providing means is arranged to provide the fine synchronization signal temporally after the coarse synchronization signal such that there is a predetermined time interval between the provision of the coarse synchronization signal and the provision of the fine synchronization signal.

28. The apparatus according to one of claims 25 to 27, wherein the providing means are configured for providing a plurality of fine synchronization signals temporally after the coarse synchronization signal.

29. The apparatus according to claim 28, wherein the providing means is configured to provide the plurality of fine synchronization signals by combining copies of the fine synchronization signals in time.

30. The apparatus according to claim 29, wherein the providing means is configured to introduce a predetermined time interval between successive copies of the fine synchronization signal.

31. Apparatus according to one of claims 25 to 30, wherein the providing means comprise a burst shaping filter for filtering the coarse synchronization signal or the fine synchronization signal.

32. The apparatus of claim 31, wherein the burst shaping filter is a roll-off filter with a roll-off factor of 1.

33. Apparatus according to one of claims 25 to 32, wherein the providing means are configured for generating a coarse synchronization signal from a data sequence and a fine synchronization signal from a further data sequence, the data sequence having a bandwidth smaller than the bandwidth of the further data sequence.

34. The apparatus of claim 33, wherein the providing means comprises a generator for generating the data sequence and for generating the further data sequence.

35. The apparatus of claim 34, wherein said generator is configured for generating said data sequence from a galois field having four elements and for generating said another data sequence from a galois field.

36. Apparatus according to claim 34 or 35, wherein the generator comprises a shift register for generating the data sequence, or for generating the further data sequence, the generator being configured to set an initial occupancy of the shift register.

37. Apparatus according to one of claims 33 to 36, in which the providing means comprise associating means for associating a complex value with each element of the data sequence to obtain a complex-valued sequence or for associating a complex value with each element of the further data sequence to obtain a further complex-valued sequence.

38. An apparatus as defined in claim 37, wherein the providing means comprises an upconverter for generating the coarse synchronization signal by upconverting the complex-valued sequence or the fine synchronization signal by upconverting the another complex-valued sequence.

39. Apparatus according to one of claims 25 to 38, wherein the providing means are configured for providing a further fine synchronization signal different from the fine synchronization signal, the providing means being configured for processing the further fine synchronization signal as the fine synchronization signal.

40. Apparatus according to any of claims 25 to 39, wherein the providing means is arranged to encode the information using a change in phase between successive values of the fine synchronisation signal in the fine synchronisation signal.

41. The apparatus of claim 40, wherein the providing means is configured to encode the information with a 180 ° phase jump between the plurality of values of the fine synchronization signal and another plurality of values of the fine synchronization signal.

42. A synchronization method for determining a position of a synchronization signal in a received signal, the synchronization signal being based on a coarse synchronization signal and a fine synchronization signal, the method comprising:

determining a part of the received signal where the fine synchronization signal is located according to the coarse synchronization signal;

determining a position of the synchronization signal in the received signal for synchronization based on the fine synchronization signal in the portion of the received signal,

wherein the step of determining the position of the synchronization signal in the received signal comprises the steps of: a fine synchronization signal in the received signal is detected,

wherein the bandwidth of the coarse synchronization signal is smaller than the bandwidth of the fine synchronization signal,

wherein the detection of the coarse synchronization signal in the received signal is performed at a first sampling rate;

wherein the detection of the fine synchronization signal is performed at a second sampling rate, an

Wherein the second sampling rate is higher than the first sampling rate.

43. A method for synchronizing a receiver and a transmitter, the transmitter configured to transmit a synchronization signal, the synchronization signal based on a coarse synchronization signal for coarse synchronization and based on a fine synchronization signal for fine synchronization, the method comprising:

sampling a received version of the synchronization signal to provide a received signal;

performing a synchronization method as defined in claim 42, to obtain a position signal indicative of a position of the synchronization signal in the received signal; and

the sampling time is controlled based on the position signal in order to synchronize the receiver and the transmitter.

44. A method for generating a synchronization signal that is transmitted to synchronize a receiver and a transmitter, the method comprising:

providing a coarse synchronization signal having a first bandwidth and a fine synchronization signal having a second bandwidth, the first bandwidth being smaller than the second bandwidth; and

the coarse synchronization signal and the fine synchronization signal are used to provide the synchronization signal.

45. A computer program having a program code for performing the method of claim 42, 43 or 44 when the program is run on a computer.

Images